武田諒

ryo-takeda's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper about 8 hours ago

Beyond Future Prediction: Denoising as Generative Adaptation for Robot Control

Paper • 2609.28339 • Published 15 days ago • 26

upvoted a paper 1 day ago

Agent Priors-guided Policy Learning

Paper • 2609.35690 • Published 10 days ago • 85

upvoted 2 papers 2 days ago

DyRAD: Radar Novel View Synthesis for Dynamic Driving Scenes

Paper • 2609.39841 • Published 8 days ago • 40

Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering

Paper • 2609.38177 • Published 9 days ago • 72

upvoted a paper 3 days ago

What Makes World Action Models Generalize? An Empirical Study of Test-Time Future Modeling

Paper • 2609.34981 • Published 9 days ago • 136

upvoted a paper 4 days ago

HybridCUA: Learning to Orchestrate GUI and CLI for Computer-Use Agents

Paper • 2609.38008 • Published 9 days ago • 47

upvoted a paper 5 days ago

Scaling Properties of Same-Family On-Policy Distillation

Paper • 2609.32722 • Published 12 days ago • 324

upvoted a paper 6 days ago

YuE2: Unifying Symbolic and Audio Music Generation at Frontier Quality

Paper • 2609.33757 • Published 11 days ago • 247

upvoted a paper 7 days ago

FoMo: Forking Moment in Generative Trajectory as a Perceptual Distance

Paper • 2609.25716 • Published 16 days ago • 24

upvoted 8 papers 8 days ago

StudentSim: Training LLM-based Student Simulators

Paper • 2609.01591 • Published Sep 1 • 496

ExplorationBench: Measuring AI Systems' Exploration in Verifiable Alien Worlds

Paper • 2609.30199 • Published 14 days ago • 30

World Action Agent: Harnessing VLMs for Robot Manipulation via World Action Rehearsal

Paper • 2609.29964 • Published 14 days ago • 15

Neural Spectral Capacity: Measuring and Designing Architectures from Network Specification Alone

Paper • 2609.23087 • Published 19 days ago • 11

Rufus-Air: An Open LLM Post-Training Recipe

Paper • 2609.29421 • Published 14 days ago • 41

Training Object Permanence in World Models

Paper • 2609.28654 • Published 15 days ago • 243

Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs

Paper • 2609.29845 • Published 14 days ago • 105

Parts-of-Speech as Emergent Categories in SAE Latent Space

Paper • 2609.29362 • Published 14 days ago • 19

upvoted a paper 9 days ago

GeoPair: Geometry-Preserving Cross-Layer Factorization for Training-Free Transformer Compression

Paper • 2609.25963 • Published 16 days ago • 17

upvoted a paper 10 days ago

Embedding Physics Priors in Robot Learning: A Survey

Paper • 2609.22319 • Published 23 days ago • 19